System and method for monitoring the operating condition of rotary electrical machines and automatic detection of mechanical and electrical faults

Abstract

The invention relates to methods for operating, maintaining and monitoring rotary electrical machines and it relates to a system and method for early detection of mechanical, electrical, load and process faults. This solution provides a more assertive detection, ease of installation, greater scalability and efficiency in preventing production losses, improving operational performance and preventing damage and risks. The method collects current and voltage signals from the machine with an acquisition module and the data goes through stages of compression, encryption, FFT, subsampling, attribute extraction, anomaly detection and fault classification with machine learning. The system comprises acquisition modules, gateway devices, a cloud processing centre and an operator interface.

Description

SYSTEM AND METHOD FOR MONITORING THE OPERATING CONDITION OF ROTATING ELECTRIC MACHINES AND AUTOMATIC DETECTION OF MECHANICAL AND ELECTRICAL FAILURES Field of invention

[0001] The present invention patent relates to methods of operation, maintenance and monitoring of rotating electrical machines, and refers to a system and a method for monitoring the operational condition and for detecting mechanical, electrical, load and process failures in rotating electrical machines. Fundamentals of the invention

[0002] A rotating electrical machine is designed to convert electrical energy into mechanical energy, or vice versa, and is basically formed by a fixed part, called a stator, and a moving part, called a rotor, which rotates along an axis.

[0003] Induction electric motors are rotating electrical machines powered by alternating current that produce torque and mechanical power from the electromechanical conversion of energy. These motors are widely used in various industries to supply mechanical power to equipment such as pumps, fans, conveyor belts, agitators, reducers, compressors, elevators, among others.

[0004] In induction electric motors, the alternating current applied to the stator windings is induced in the rotor by the transformer action between the coils of these parts. The current induced in the rotor generates a magnetic flux, which seeks to align itself with the rotating magnetic field of the stator, causing rotational movement in the rotor. The rotor shaft is connected to a load by means of a coupling.

[0005] On the other hand, conversely, in induction generators, a primary mechanical machine promotes the rotation of the rotor. Thus, in the presence of electrical energy residual, the rotor coils induce alternating current in the stator windings through transformer action, generating electrical energy.

[0006] Since rotating electrical machines have moving parts in contact with fixed parts, friction occasionally occurs between the rotating part and static parts. For this reason, the machine experiences dynamic and frictional forces, often causing conditions conducive to mechanical failures, such as unbalance, misalignment, wear, loosening of the housing or moving components, corrosion, blockage, obstruction, heating and other phenomena that can cause failures.

[0007] Rotating electrical machines are also subject to a variety of electrical faults. These include phase imbalance, voltage transients, harmonic distortion, sigma currents, and overload. Phase imbalance can cause overheating and melting of conductors, and is often the result of design flaws and the addition of unforeseen loads. Voltage transients are surges from external sources and oscillations caused by load maneuvers and capacitor bank switching. This transient effect can cause insulation breakdown and failures in the medium and long term. Harmonic distortion results from capacitive and inductive components in the line, and from electronic loads that have diodes and transistors and generate discontinuity in the electrical current. Such harmonics are energy losses that, over time, imply a loss of motor efficiency and deterioration of insulation.Sigma currents are parasitic currents present in the motor conductors and can cause loss of efficiency and shorten the motor's useful life. To avoid them, the conductors must be properly dimensioned and the connections well made. Finally, overload occurs when a motor is operating above its nominal torque and current, generating overheating and reducing its useful life. Frequent electrical faults contribute to the deterioration of parts of the rotating electrical machine, causing failures such as broken rotor bars and damage and burning of the stator windings.

[0008] Thus, in an industrial plant in which a conglomerate of rotating electrical machines are installed and operating, mechanical and electrical failures and, consequently, unscheduled production stoppages, represent losses. significant financial consequences for companies, as they often involve loss of production, high repair costs, the need to purchase spare parts or a new machine, in addition to causing potential risks to operators exposed to the failure. In this scenario, well-executed preventive and predictive maintenance is essential to ensure asset availability, production efficiency and safety for operators and maintainers. Typically, preventive and predictive monitoring services for mechanical and electrical problems in rotating electrical machines are performed periodically and use vibration analysis, thermographic inspection and electrical signature analysis (ESA).

[0009] Vibration analysis is a traditional inspection method widely used in industry for detecting mechanical faults in rotating electrical machines, as it offers good reliability in fault diagnosis. This method exploits the fact that each type of mechanical fault generates a specific vibration pattern in the machine, thus generating a characteristic signature of such behavior. Thus, from the analysis of the spectrogram of the vibration signal, the condition of the machine can be inferred.

[0010] Despite being a widely adopted strategy, vibration analysis requires that the sensor, basically consisting of a three-axis accelerometer, be installed directly on the asset to be monitored, usually on the bearings of the rotating electrical machine, to read the vibration signals. However, such assets are often installed in hard-to-reach, unhealthy and dangerous locations, thus placing the operator who will install the sensor at occupational risk. Furthermore, the sensors are continually subject to adverse operating conditions, such as high and low temperatures, dust, process residues, high vibrations, mechanical impacts, among others, which can cause reading errors and physical damage to the equipment, leading to a reduction in its useful life.

[0011] Thermographic analysis of rotating electrical machines allows for the precise identification of thermal faults and wear at specific points on the machine. This means that any damage caused by excessive heat can be detected without having to disassemble the motor. However, this method requires trained professionals, usually using a digital thermal imager, to travel to the location of the machine. installation of the equipment to perform the inspection. Therefore, it is not a method with adequate practicality to monitor assets in difficult to access, unhealthy locations, or any location that puts the inspector at occupational risk.

[0012] ESA is a method in which current and voltage signals are captured and analyzed based on signature patterns displayed on the spectrogram. This analysis allows the detection of patterns associated with failures and the identification of the part of the equipment in which the failure is occurring. An important advantage of this method in relation to vibration and thermography analysis is that the test can be performed remotely, in the electrical panel, for example, without the need to install sensors directly on the asset. Another important benefit is that this technique provides motor energy analysis, offering more information about the equipment. Regarding the disadvantages, the ESA technique depends on sophisticated equipment, generally expensive, and which often does not have the desired characteristics of high sampling rate and resolution.

[0013] The maintenance methods described above depend on services performed by trained and experienced specialists and on robust, high-precision equipment, which represents a significant portion of the operating costs of an industrial plant. Although such methods are accurate and reliable, early-stage problems often go undetected because they are imperceptible to a human operator and are usually performed only periodically, not continuously. Therefore, a failure in the development phase, still with low severity, when undetected, can quickly evolve into a high-severity failure, which can lead to equipment and process shutdowns, in addition to causing damage to machine components and even other coupled parts.

[0014] In order to overcome the difficulties presented, newer systems and methods use tools to continuously monitor variables in rotating electrical machines in order to check the condition of the assets more accurately and earlier. These systems include sensors installed permanently on the equipment, or for a long period of time, connected to a local network or a mobile network, and advanced Digital Processing techniques. of Signals (PDS) to infer the condition of machines from the collected signals, and based on reference standards.

[0015] Regarding vibration analysis, there are currently solutions that collect and process data continuously and transfer it to a local or cloud server, where an application based on PDS and Data Science techniques, which may include the application of models based on Machine Learning, performs the analysis of the machine's condition and the diagnosis of possible failures still at an early stage. In this case, the need for periodic inspections within a maintenance plan by an experienced professional is reduced, since data collection is performed continuously and uninterruptedly, and the specialist application, based on data and machine learning, provides failure prediction and optimization of maintenance management.

[0016] However, even with the use of continuous monitoring combined with the use of machine learning methods for fault diagnosis, vibration analysis still has significant disadvantages compared to monitoring with electrical signal analysis. First, this method does not allow for the direct detection of electrical faults, and depends on the appearance of mechanical symptoms resulting from the primary problems to identify the fault. Furthermore, this method does not offer analysis of power quality and operational performance, an essential tool for reducing energy consumption and optimizing machine performance. Another disadvantage is the need to install a sensor on each machine in a set of coupled equipment, increasing the cost of line maintenance.Furthermore, in vibration analysis, the location and orientation of the sensor determines the quality of the collected signals, making the installation totally dependent on the conditions of the asset, the environment and the structural composition of the industrial plant.

[0017] Considering the problems encountered in the maintenance practices of rotating electrical machines highlighted and discussed above, a system to detect failures early in a more assertive manner, with greater practicality in installation, with greater scalability, and that offers additional analyses to optimize the operation, with the aim of avoiding production losses, improve operational performance and prevent damage to equipment and risks to operators.

[0018] In the present invention, a system and a method for monitoring the operational condition of rotating electrical machines and detecting mechanical and electrical faults automatically were developed. The developed method continuously collects electrical current and voltage signals that supply the machine, at a high sampling rate and high resolution, from a data acquisition module and transducers installed in the electrical panel. The collected, treated and pre-processed electrical current and voltage signals are applied to intelligent models based on machine learning, on cloud servers for fault identification and classification.

[0019] The method developed in this invention can detect mechanical and electrical problems in rotating electrical machines at an early stage. In the presence of faults, vibrations and anomalies arise that affect the air gap between the machine's stator and rotor, causing disturbances in the magnetic field. Such disturbances affect the behavior of the current and electrical voltage, and insert patterns associated with the faults, which can be identified at an early stage by the machine learning-based method developed. In addition, electrical faults are detected in a much more assertive and efficient manner compared to other methods, since they directly affect the motor's magnetic field and, consequently, the current and electrical voltage, which the developed system continuously monitors.

[0020] The system and method developed offer a series of advantages over traditional processes and even those most recently adopted by the industry. Installing the data acquisition module in the electrical panel offers greater practicality, agility and safety in the installation of the system. In addition, it facilitates the installation of the monitoring system in places that are difficult to access and highly unhealthy and dangerous. Furthermore, the transducers are not exposed to adverse conditions typical of places with extremely high temperatures or submerged installations.

[0021] Another advantage of the developed method offered by the installation of the data acquisition module in the electrical panel is the ability to monitor a series of coupled machines using only one monitoring device. For example, a motor-reducer-agitator set can be monitored by applying the method to the electrical signals that feed the line, without the need to install a sensor in each machine, since the torque generated by the electromechanical energy conversion in these machines is derived from the same electrical energy source. In addition, the developed method can identify the machine in a fault condition and the type of fault, since each machine-fault set presents a specific electrical signature in the collected signals.

[0022] The construction of a database of electrical signals collected from rotating electrical machines with a wide variety of operating conditions and fault conditions, together with the use of machine learning techniques, allows the construction of a robust, accurate and efficient model to diagnose faults at an early stage. During operation, continuous data collection makes the developed method even more assertive, since the machine learning-based model can be improved with new data continuously generated by the monitored machine.

[0023] In addition to offering continuous monitoring with early diagnosis of faults in electrical machines, the present invention also provides power quality analysis and machine performance metrics, allowing operators to make more assertive, data-based decisions, resulting in more efficient operations in terms of production and energy consumption. Brief description of the drawings

[0024] The figures and flowcharts contained in this patent application are briefly described below: Figure 1 presents an illustrative example of the overview of the proposed system, according to a main realization. Figure 2 illustrates an example of the overview of the proposed system, according to a secondary realization. Figure 3 shows the installation of the data acquisition module in the electrical panel, according to an implementation. Figure 4 reveals the data acquisition module and its components, according to an embodiment. Figure 5 presents the flowchart of the data collection and processing process locally, according to an implementation. Figure 6 shows the flowchart of data processing steps in the processing center on the cloud computing platform according to an embodiment. Figure 7 shows the flowchart of the training and application process of machine learning models, according to an implementation. Figure 8 presents the flowchart of the fault detection and operator validation process, according to an implementation. Description of the invention

[0025] The present invention, called “System and method for monitoring the operational condition of rotating electrical machines and automatic detection of mechanical and electrical faults”, comprises a system and a method for continuous monitoring of electrical current and voltage of industrial rotating electrical machines from the electrical panel, and diagnosis of an extensive list of faults originating from the machine itself, the driven load, the process, or other elements coupled to the line, even while still in the development stage.

[0026] The proposed system and method, together, offer a way to detect failures early in a more assertive manner, with greater practicality in installation, with greater scalability than state-of-the-art methods, and are more efficient to avoid production losses, improve operational performance and prevent damage to equipment and risks to operators.

[0027] The monitored rotating electrical machines can be, without limitation, induction electric motors, generators, pumps, fans, conveyor belts, agitators, reducers, elevators, turbines, compressors, gearboxes.

[0028] Among the faults detected by the proposed method are, without limitations, wear, scratches and bearing failures, bearing failure, eccentricities, broken bars in the rotor, shaft misalignment, shaft imbalance, shaft clearance, soft foot, overload, voltage transients, phase imbalance, harmonic distortion, Sigma currents, pump cavitation, pump clogging, corrosion of parts, resonance, seal assembly failure, leakage, rotary looseness, structural looseness, operator failure.

[0029] In summary, the developed method continuously collects electrical current and voltage signals that power a rotating electric machine, at a high sampling rate and high resolution, from a data acquisition module, current transformers, and voltage transformers, installed on the electric panel. The collected signals are then processed in a processing center, where machine learning models are applied for fault classification.

[0030] Figure 1 presents an illustrative example of the general view of the proposed system, according to a main embodiment. In this main embodiment, the proposed system is formed by N data acquisition modules (101) and M gateway modules (102) in the industrial plant, a processing center (103) in a cloud computing platform (104), and an operator interface (105). In this embodiment, each gateway device (102) communicates with a group of n data acquisition modules (101) through a local network (106), and with the processing center (103) through the internet network (107). The number N and the number M vary, respectively, according to the number of data acquisition modules (101) and gateway devices (102) installed in the industrial plant. The number n varies according to network and signal quality constraints, such as distance, physical barriers and electromagnetic emissions between devices and in the environment in which they are located, in addition to hardware limitations. In the illustrative example in Figure 1, N is equal to 6, M is equal to 2, and n is equal to 3.

[0031] Figure 2 presents an illustrative example of the general view of the proposed system, according to a secondary embodiment. In this secondary embodiment, the proposed system is formed by N data acquisition modules (101), individually coupled to N gateway devices (102) in the industrial plant, a processing center (103) in a cloud computing platform (104), and an operator interface (105). The gateway device (102) communicates with the processing center (103) through the internet network (107). The number N varies according to the number of data acquisition modules (101) installed in the industrial plant. In the illustrative example of Figure 2, N is equal to 4.

[0032] Figure 3 shows the installation of a data acquisition module (101) adapted to monitor a three-phase rotating electrical machine (301), according to a preferred embodiment. In this embodiment, the data acquisition module (101) is installed inside the electrical panel (305) in which there is a machine activation circuit (304). The machine activation circuit (304) receives three-phase electrical power from the electrical network input (306).

[0033] In one embodiment, current transformers (CTs) (302) are installed in each phase conductor of the machine activation circuit (304) that powers the rotating electrical machine (301). The CTs can be of the split-core or closed-core type, or Rogowski coils, any of these being of various types, categories and sizes, depending on the intensity of the current to be measured.

[0034] In one embodiment, branch points on the conductors (303) are made on each phase conductor to derive electrical voltage information to the data acquisition module (101). The branch points on the conductors (303) can be made from the output connections of the machine drive circuit (304), or from connections made directly on the power conductors of the rotating electrical machine (301).

[0035] Figure 4 shows the data acquisition module (101) according to a preferred embodiment. In this preferred embodiment, the data acquisition module (101) is composed of a microcontroller (401), an analog-to-digital converter (402), a set of three potential transformers (PTs) (403), a memory (404), a transceiver (405), and a direct current (DC) power supply (406).

[0036] In one embodiment, the microcontroller is responsible for executing the data collection control steps and communicating with the other components of the data acquisition module (101).

[0037] In one embodiment, the set of VTs (403) receives the electrical voltage signals derived from the branch points in the conductors (303) of each phase conductor feeding the rotating electrical machine (301).

[0038] In one embodiment, the analog-to-digital converter (402) receives as input the analog signals captured from the secondary windings of the CTs (302) and the VTs (403), and converts them into digital signals, which are transferred to the microcontroller (401).

[0039] In one embodiment, the memory (404) stores data collected by the data acquisition module (101).

[0040] In one embodiment, the transceiver (405) enables communication and performs data transfer between the data acquisition module (101) and the gateway device (102), and can be implemented in three configurations. In a first configuration, the transceiver (405) is a Wi-Fi communication module, and allows the wireless connection of the data acquisition module (101) to the local network (106) to communicate with the gateway device (102). In a second configuration, the transceiver (405) is an Ethernet communication module, and enables the connection of the data acquisition module (101) to the local network (106), through a network cable, to communicate with a gateway device (102). In a third configuration, the transceiver (405) is an adapter to couple the gateway device (102) physically to the data communication module (101), and provide communication between them.

[0041] In one embodiment, the DC power supply (406) is responsible for providing adequate electrical power to all components of the data acquisition module (101). The DC power supply (406) has as its power source the electrical grid input (306), available in the electrical panel (305), in which the data acquisition module (101) is installed.

[0042] The processing center (103) is made up of various computing services and tools available on the cloud computing platform (104), and is accessed and operated in the cloud via the internet network (107). The services that make up the processing center include data ingestion, transformation, storage, processing and analysis, as well as training and execution of machine learning models.

[0043] Figure 5 shows the flowchart of the data collection, processing and transfer process to the processing center (103), according to one embodiment. The input of this process are the electrical current and voltage signals (505) that feed the rotating electrical machine (301). The data acquisition module (101) performs the data collection step (510) and transfers the data to the gateway device (102). This, in turn, performs the data compression step (515), data encryption (520), and data transfer to the cloud processing center (525). At the end of this process, the result is the compressed and encrypted data available in the processing center (530).

[0044] Figure 6 shows the flowchart of data processing in the processing center (103), according to one embodiment. The input of this process is the compressed and encrypted data available in the processing center (530), transmitted by the gateway device (102). In step 605, the processing center decodes the data. Next, it executes the step of applying the fast Fourier transform (FFT) (610) to obtain the representation in the frequency domain. Then, it performs the step of subsampling the signals in the time domain (615), and the step of extracting attributes in the time and frequency domain (620).

[0045] The feature extraction step of the time and frequency domain signals (620) includes one or more feature extraction and selection techniques, without limitations, such as mean, median, standard deviation, variance, kurtosis, skewness, covariance matrix, Fourier coefficients, natural frequency, fundamental frequencies, frequency harmonics.

[0046] In step 625, the processing center (103) performs the anomaly detection step in the signals with one or more statistical techniques and one or more machine learning techniques, using the data and attributes obtained up to the previous step.

[0047] In a secondary embodiment, step 625 may include principal component analysis (PCA) technique and an autoencoder neural network.

[0048] In the event of an undetected anomaly, the result of this process is the indication of the machine in normal condition (635).

[0049] In the event of an anomaly being detected, the processing center (103) performs the step of identifying the machine's operating condition and classifying the fault with one or more machine learning techniques (640), using the data and attributes obtained up to step 620.

[0050] In a secondary embodiment, step 640 may include convolutional neural networks and one or more decision tree-based boosting algorithms.

[0051] In step 645, the result will be an indication of the failure detected in the previous step.

[0052] Machine learning techniques applied in steps 625 and 640 may include other algorithms, without limitation, such as Artificial Neural Networks, Recurrent Neural Networks, Decision Trees, Random Forest, extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), CatBoost k-Nearest Neighbors (KNN), Naive Bayes and ensemble classifiers.

[0053] Figure 7 shows the flow of the training process and application of machine learning models, according to an implementation. In this implementation, the training stage of the machine learning models is initially performed. machine learning (710), which are trained, optimized and evaluated with a database with a plurality of electrical current and voltage signals and their attributes extracted in the time and frequency domain (705). When in production, the application of the machine learning models (715) is executed to the new collected electrical current and voltage signals and their attributes extracted in the time and frequency domain (720). The new collected signals (720) are also stored in the processing center (103) and inserted into the database (705) in order to be used in a new training process of the machine learning models (710) for improvement and specialization of the machine learning models. At the end of the process, step 725 is executed, indicating the operational condition of the machine and the associated failure.

[0054] The database with a plurality of electrical current and voltage signals and their extracted attributes in the time and frequency domain (705) encompasses a variety of data collected and attributes extracted from electrical current and voltage signals of a plurality of rotating electrical machines (301), of diverse types, with different loads, in varied operating conditions, and under diverse mechanical, electrical, load and process fault conditions.

[0055] Figure 8 shows the flowchart of the fault detection and operator validation process, according to an implementation. In this implementation, the machine learning models (805) are initially trained. In step 810, the machine learning model detects an anomaly, and in step 815, the machine learning model classifies the identified fault. As a result, the step of indicating the machine's operating condition and the associated fault (820) is executed. Next, the operator alert step (825) is performed. The operator then receives the notification and evaluates the machine's condition to execute step 830, operator validation. Based on operator validation, the machine learning models are trained with new data collected and with the information on the machine's operating condition validated or rejected by the operator.This procedure contributes to the improvement and specialization of the model in the rotating electrical machine (301) that is being monitored.

Claims

CLAIMS 1. Method for monitoring the operational condition of rotating electrical machines and automatically detecting mechanical and electrical faults, characterized by comprising the steps of: a. collecting data (510) of electrical current and voltage signals (505) from a rotating electrical machine (301) with a data acquisition module (101); b. transferring the data collected by the data acquisition module (101) to the gateway device (102); c. in the gateway device (102), performing the steps of: i. data compression (515); ii. data encryption (520); and iii. transferring data to the cloud processing center (525) via the internet network (107); d. in the processing center (103), performing the steps of: i. decoding the data (605); ii. applying the fast Fourier transform (FFT) (610) to obtain the representation in the frequency domain; iii. time domain subsampling (615); and iv.extraction of attributes from signals in the time and frequency domain (620);. e. in the processing center (103), perform the anomaly detection step in the signals with one or more statistical techniques and one or more machine learning techniques (625); f. in the processing center (103), in the event of an anomaly detected, perform the step of identifying the machine's operating condition and fault classification with one or more machine learning techniques (630).

2. Method, according to claim 1, characterized in that the data acquisition module (101) is installed inside the electrical panel (305), and collects electrical current and voltage signals (505) from a rotating electrical machine (301) with current transformers (CTs) (302) and branch points in the conductors (303) installed together with the phase conductors, also inside the electrical panel (305).

3. Method according to claim 1, characterized in that the step of detecting anomalies in the signals with one or more statistical techniques and one or more machine learning techniques (625) includes the technique called principal component analysis (PCA) and includes an autoencoder neural network.

4. Method according to claim 1, characterized in that the step of identifying the machine operating condition and fault classification with one or more machine learning techniques (630) includes convolutional neural networks and one or more boosting algorithms based on decision trees.

5. Method, according to claim 1, characterized by obtaining validation from the operator (830) on the detected failures, and inserting the received validation, together with new data collected from the failing machine, into the training process of the machine learning models (805).

6. Method according to claim 1, characterized in that the data transfer between the data acquisition module (101) and the gateway device (102) occurs remotely through a local wireless or wired communication network (106).

7. Method according to claim 1, characterized in that the rotating electric machine is an induction electric motor, a generator, a pump, a fan, a conveyor belt, an agitator, a reducer, an elevator, a turbine, a compressor, a gearbox.

8. The method of claim 1, wherein the machine failure operating condition includes wear, bearing scratches and failures, bearing failure, eccentricities, broken rotor bars, shaft misalignment, shaft imbalance, shaft play, soft foot, overload, voltage transients, phase imbalance, harmonic distortion, Sigma currents, pump cavitation, pump clogging, parts corrosion, resonance, seal assembly failure, leakage, rotary looseness, structural looseness, operator failure.

9. System for monitoring the operational condition of rotating electrical machines and automatically detecting mechanical and electrical faults, characterized by comprising: a. a data acquisition module (101) for collecting electrical current and voltage signals and transferring the electrical current and voltage signal data to the gateway device (102); b. a gateway device (102) for compressing the electrical current and voltage data, encrypting the electrical current and voltage data, and transferring the electrical current and voltage data to the processing center (103); c. a processing center (103) on a cloud computing platform (104): i. in which the electrical current and voltage data are decoded; ii. in which the FFT is performed; iii. in which the electrical current and voltage data are subsampled in the time domain; iv. in which the attributes of the signals are extracted in the time and frequency domain; v. in which one or more statistical techniques and one or more machine learning techniques are implemented for detecting anomalies in the electrical current and voltage signals; vi. in which one or more machine learning techniques are implemented for identifying the operating condition of the machine and classifying faults.

10. System, according to claim 9, characterized in that the data acquisition module (101) is installed in the electrical panel (305) and collects electrical current and voltage data from a rotating electrical machine (301) with CTs (303) and TPs (304) installed together with the phase conductors that supply the rotating electrical machine (301).

11. System, according to claim 9, characterized in that the processing center (103) is adapted to perform principal component analysis (PCA) and an autoencoder neural network in the anomaly detection step in the signals with one or more statistical techniques and one or more machine learning techniques (525).

12. System, according to claim 9, characterized in that the processing center (103) is adapted to apply convolutional neural networks and one or more boosting algorithms based on decision trees in the stage of identifying the machine's operational condition and classifying failure with one or more machine learning techniques (530).

13. System, according to claim 9, characterized in that the processing center is adapted to obtain validation from the operator (830) about the failures. detected, and insert the received validation, together with new data collected from the failing machine, into the training process of the machine learning models (805).

14. System, according to claim 9, characterized in that the gateway device (102) communicates remotely with the data acquisition module (101) through a local wireless or wired communication network (106).

15. System according to claim 9, characterized in that the rotating electric machine (301) is an induction electric motor, a generator, a pump, a fan, a conveyor belt, an agitator, a reducer, an elevator, a turbine, a compressor, a gearbox.

16. The system of claim 9, wherein the machine failure operating condition includes wear, bearing scratches and failures, bearing failure, eccentricities, broken rotor bars, shaft misalignment, shaft imbalance, shaft play, soft foot, overload, voltage transients, phase imbalance, harmonic distortion, Sigma currents, pump cavitation, pump clogging, parts corrosion, resonance, seal assembly failure, leakage, rotary looseness, structural looseness, operator failure.

Images